Fig. 3
From: Predicting weighted unobserved nodes in a regulatory network using answer set programming
MajS workflow. In light green, we show the input data: the interaction graph, IG, and the discrete observation list, Obs. Then, we apply the logical rules implemented in MajS. We test the consistency and in case of inconsistency, we add artificial influences using K as a fixed parameter. That way we obtain answer sets that respect the logical rules. We minimise the artificial influences added to these answer sets and obtain an optimal subset of them. Finally, we project the optimal answer sets to obtain as output the predicted nodes of our model. clasp is the Answer Set Programming solver [17] used to implement most of MajS steps. Only the projection step was implemented in Python